79trust / 100

Guard Check

by Claw-in-a-Box in Email & messaging

x402 APIPassing, checked 5 min ago

Spend-policy verdict for AI agents: POST a proposed spend, get allow/review/deny with the exact rules that fired (per-tx caps, per-agent daily ledger, destination allowlists, time windows) - 'review' can page a human on Telegram for approval before the agent proceeds. Pay per call.

POST https://api.clawinabox.xyz/paid/v1/guard/check

Last 30 days

All checks passedSome failedAll failedNot checked
Uptime
100%
Response time
447 ms typical, 666 ms slowest 5%
Last check
5 min ago
Next check
in 27 min

How to call it

# See the payment challenge (nothing is charged)
curl -i -X POST "https://api.clawinabox.xyz/paid/v1/guard/check" \
  -H "content-type: application/json" \
  -d '{"agent_id":"agent-7","amount":150,"destination":"merchant-x","policy":"standard"}'
import { wrapFetchWithPayment } from "@x402/fetch";
import { x402Client } from "@x402/core/client";
import { ExactEvmScheme } from "@x402/evm/exact/client";
import { privateKeyToAccount } from "viem/accounts";

const client = new x402Client().register(
  "eip155:8453",
  new ExactEvmScheme(privateKeyToAccount(process.env.AGENT_KEY)),
);
const pay = wrapFetchWithPayment(fetch, client);

// Not sure it's safe to pay? Preflight it first for $0.005:
// GET https://toolvet.app/api/v1/check?url=https%3A%2F%2Fapi.clawinabox.xyz%2Fpaid%2Fv1%2Fguard%2Fcheck
const res = await pay("https://api.clawinabox.xyz/paid/v1/guard/check", {
  method: "POST",
  headers: { "content-type": "application/json" },
  body: JSON.stringify({"agent_id":"agent-7","amount":150,"destination":"merchant-x","policy":"standard"}),
});
console.log(await res.json());

Example input

{
  "agent_id": "agent-7",
  "amount": 150,
  "destination": "merchant-x",
  "policy": "standard"
}

Example output

{
  "approval_id": "apr_9f2c1a",
  "poll": "/v1/approvals/apr_9f2c1a",
  "reasons": [
    "amount 150 exceeds review threshold 100"
  ],
  "triggered": [
    "review_threshold"
  ],
  "verdict": "review"
}

Security scan

  • No findings. We scan names, descriptions and tool definitions for hidden instructions and other prompt-injection patterns.

Recent checks

WhenResultHTTPTimePrice
5 min agoPassed402447 ms$0.01
4 h agoPassed402358 ms$0.01
7 h agoPassed402470 ms$0.01
12 h agoPassed402343 ms$0.01
15 h agoPassed402670 ms$0.01
18 h agoPassed402352 ms$0.01
23 h agoPassed402666 ms$0.01